Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Adventure game graphics with DALL-E 2

hpjansson.org

Post reply on HN
Print view View original

81–90 of 162 posts

Re: Adventure game graphics with DALL-E 2

#81

Post by shreyshnaccount »

awb wrote: post #68
jillesvangurp wrote: post #38

Earlier quoted context omitted.

We're at the beginning of a rapidly evolving situation where out of the blue it is now possible to direct an AI to produce imagery that is interesting using just language. From there to moving animations is a relatively small step and already being worked on. From there to photo realistic scenes directed by just words not that big of a leap either and also something that is being worked on. Are there constraints, gli…

> From there to moving animations is a relatively small step and already being worked on. From there to photo realistic scenes directed by just words not that big of a leap either and also something that is being worked on. I don’t think DALL-E can even persist a single 2d object between scenes. For example, if you create an animated character you like, I don’t think it’s possible to ask DALL-E to render that same ch…

I think you can. you need to fine tune it and make it learn about a character. I've seen prompts about Darth Vader and the beans from among us produce decent outputs, so Dall E can totally learn how a specific character looks as long as it's in its training set atleast

Re: Adventure game graphics with DALL-E 2

#82

Post by simion314 »

skytrue wrote: post #26

I started off feeling hopeful/excited by OpenAI, GPT-3, DALL-E, etc, as a creator/writer. As time has progressed and I've used these tools and other machine learning tools to create media, I've become disillusioned. It feels as if this is about to explode an existing problem, namely, there's too much content in the world already. Lowering the effort bar for content creation and generation is, in some ways, going to c…

> Lowering the effort bar for content creation and generation is, in some ways, going to cause a deluge of low-quality/mid-quality content to be generated.

This is not a problem, the issue is with Steam of whoever curates this content for us, with some good filters you can avoid the problem. My first program and game was garbage but it was not on GitHub or Steam to bother anyone, so the issue is not that people have access to good tools (I was using some Visual Studio for students edition) the issue might be that maybe some services have bad filters.

For new indie devs, please use whatever tools make your vision , publish them, ignore the haters, you will probably not make a living from Patreon or donations but if you love what you doa nd makes you happy that is the important thing.

Re: Adventure game graphics with DALL-E 2

#83

Post by aembleton »

GrumpyNl wrote: post #75
thorum wrote: post #47

Earlier quoted context omitted.

One cool thing about Stable Diffusion (the current leading competitor to DALLE) is that you can create images in a similar style by configuring the seed number. Different prompts with the same seed number can be used to create different images with similar visual styles, character design, etc. For example, this person used SD to generate portraits of the same woman at different ages, from an infant to an 80 year old:…

Why does it have so many problems with the eyes?

Maybe the uncanny valley for eyes are deeper. Our brains might have evolved to notice eyes more.

Just speculating though, there might be a technical reason too.

Re: Adventure game graphics with DALL-E 2

#84

Post by CamperBob2 »

darishante wrote: post #62
pugworthy wrote: post #43

Why not feed all the descriptions from say, original Colossal Cave Adventure game into DALL-E 2? "YOU ARE STANDING AT THE END OF A ROAD BEFORE A SMALL BRICK BUILDING. AROUND YOU IS A FOREST. A SMALL STREAM FLOWS OUT OF THE BUILDING AND DOWN A GULLY."

The first thing I thought when reading this was "wow, someone must have fed Infocom games' location descriptions into an AI". Would still like to see that.

So frustrating that there's no way to access this tech without putting your name on a list and waiting an unspecified amount of time (months?). It's the sort of thing many people would pay well for, even at its current stage of development.

Re: Adventure game graphics with DALL-E 2

#86

Post by bigfudge »

ALittleLight wrote: post #64
NegatioN wrote: post #56

Earlier quoted context omitted.

The quality of the upper echelon of art may be raised, but there's still a discovery problem there. Having to sift through stuff to find the gems is already an issue imo. The OP makes a decent (pessimistic) point

I think AI will be pretty good at doing recommendations. Show you a bit of random, get your likes and dislikes, exploit what the algorithm learned. TikTok does this well already and, I expect, will continue to do well when the content is AI generated.

Tiktok optimises for what people spend a long time looking at, but I don’t think that anyone would claim the metric it uses is what we would want to define as quality in the broader sense.

Re: Adventure game graphics with DALL-E 2

#87

Post by jstanley »

> A common defense claims that the model learns from the training set the same way a human student would, implying human rules (presumably with human exceptions) should apply to its output. This can seem like a reasonable argument in passing, but besides being plain wrong, it's too facile since DALL-E is not human-like.

This is nowhere near enough to refute the idea that human rules on fair use should also apply to works you create using DALL-E.

The fact that DALL-E can't own the output is irrelevant. Firstly, the human operator can quite happily claim ownership of the output. Secondly, I don't see how ownership of the output is even related to whether the training data was used fairly.

Re: Adventure game graphics with DALL-E 2

#88

Post by NegatioN »

ALittleLight wrote: post #64
NegatioN wrote: post #56

Earlier quoted context omitted.

The quality of the upper echelon of art may be raised, but there's still a discovery problem there. Having to sift through stuff to find the gems is already an issue imo. The OP makes a decent (pessimistic) point

I think AI will be pretty good at doing recommendations. Show you a bit of random, get your likes and dislikes, exploit what the algorithm learned. TikTok does this well already and, I expect, will continue to do well when the content is AI generated.

I work in the area of recommendations, and this is not a solved problem at all. You can only recommend what has been shown (without doing coldstart). One major issue is that other forms of content than 30sec clips can't easily utilize TikToks way of bootstrapping engagement when the item is fresh. Not everyone will understand or appreciate a "new Shakespeare" and it may fall by the wayside.

I too hope it gets better, but it's hard to replace a panel of experts that have sifted through their subject when it comes to quality recommendations in some fields.

Re: Adventure game graphics with DALL-E 2

#90

Post by 411111111111111 »

flohofwoe wrote: post #78
jillesvangurp wrote: post #38

Earlier quoted context omitted.

We're at the beginning of a rapidly evolving situation where out of the blue it is now possible to direct an AI to produce imagery that is interesting using just language. From there to moving animations is a relatively small step and already being worked on. From there to photo realistic scenes directed by just words not that big of a leap either and also something that is being worked on. Are there constraints, gli…

I wonder how tweakable those generated images are. The most important requirement for 'game art creation' is that it is created in a tight feedback loop between an artists, an art director who needs to enforce an overall artistic vision, and the cold, hard requirements of the game design. Can I tell the AI "that looks great, keep the trees and house, but I need the door on the right side, and the sky a bit less cloud…

It's going to dominate on short form media platforms such as TikTok once there is an a beta version I think.
Post reply on HN